Instead of waiting for data to appear and using statistics and rule mining to analyze it, computational scientists can work with domain scientists to propose a small but nicely-covering sample of experiments, evaluate the results and propose a refined set. The underlying technology includes two-factor combinatorial design, circuit inference, and fast algorithms for time series analysis. This talk discusses these tools in the context of a series of case study collaborations with biologists and physicists.

Joint work with Gloria Coruzzi, Allen Mincer, and Yunyue Zhu.